I have noticed over the last few months that there were sourceforge projects I showed an interest in that were binary only. For a time I tried to convince myself the project maintainers were just too busy and they were behind posting the sources. Now I see that we could rename the site, Binary-Forge. I remember clearly that for a long time, SF was the premier site for hosting open source development. I myself developed an open source XML project on SF (xmltools). There is a big difference though between hosting adware, and malware. Personally I had more respect for the mode of operation where the sources were posted, with the appropriate cryptographic hashes, and a polite warning that prepared binaries were included as a courtesy, but the build-it-yourself from sources technique was safer, primarily because you could review every line of the code if desired. Just what made SF the premiere site were the obvious superior features, but topping it off of course was the reputation they earned in the early days. Should someone with the time and resources wish to compete with them now, the hosting meme is very well defined and with SF reputation currently encumbered as it is, they could easily be replaced in the wink of an eye. It makes me sad to see SF come to this, and I can only assume it has happened because the people who had the image in mind of a trustworthy hosting site are now gone, and bean-counters have taken over. It's too bad that short-sightedness makes aberrant practices seem more profitable than retaining the trust and love of the developer community.
Coming from the world of proprietary software, I was slow to accept open source. Given time though I became a fan of the GNU universe. From the start though I found Richard hard to take. In time Linux became strong enough to compete with OpenBSD and there the contest was between the egos of those two. This free enterprise-like competitive environment, both in systems and applications has resulted in a very rich choice of software available for us to use, free or otherwise. As far as the dynamics within development organizations, this is similar to the climate in early silicon valley where people with a better idea (or so they thought) would strike out on their own (we would call it forking a project). Some people lack social skills, or the balance between computing skills and social skills. Just because someone had a great idea and the computing skills to create a widely used program, doesn't mean they can exercise shared dominion over the project when it grows up. John draper is a good example of this. Surely a man with moments of genius, but in my opinion, no-one who should ever be managing others. Two factors seem important to me in the growth of open source projects. One is the emotional chemistry that bonds people together such that they are happy to donate time towards the project's goals. The other is the financial aspect, because some projects have financial needs to support healthy growth, and people with money have to relate to the project leaders if money can be expected to flow. There are people that wear suits and have technical skills, and they are very valuable to open source projects as well.
See my blog
I couldn't agree more. Linksys was all right in the ancient past, But beyond the physical aspects of their products, they are erratic when it comes to agreements made by their tech support. I bought a pair of their top of the line 802.11n routers for a project. Design requirements included the gb switch. Within hours it became apparent the routers we're either flakey (hardware) or the firmware was terminally buggy. After extensive phone support testing, they agreed to RMA and replace the routers. I paid to return the two routers and several weeks later a bright and shiny new pair arrived. Unfortunately the new routers were just as unstable as the original pair. The routers were v1.0 and I noticed on the web site that they also had a v1.1 version. After another hour on the phone with their tech support people, they agreed to RMA the latest pair and ship me two v1.1 units. I paid again tho return the pair of routers. Several weeks later I get a another shipment from them and inside a find two bright and shiny v1.0 units. I checked the shipper and it indicated that I was to be shipped the v1.1 units. But someone had crossed out the "v1.1" and penciled in "v1.0". I was astounded, and subsequently I have not purchased any more Linksys products.
I agree that the systems are hopelessly complex. I have observed that the average Windows installation consists of about a million files. With that magnitude of filenames, it is unrealistic to expect anyone to know what each one is and what it does, even of the names are somewhat mnemonic. When you have a million files (1,000,000), each one is approximately 0.000001% of the total. That sort of perfection is difficult to achieve, and it doesn't happen without a NASA-like effort. I have worked for employers that expected me to watch every line of code execute in a debugger before considering it committable. If each file consisted of at least a thousand lines of code, that would mean there are a billion lines of code in the system. Ouch!!
Somewhere around 1984, Intel released their 80286, which had special features for operating system writers, including boundary protection implemented through memory descriptors. It appears that Microsoft made a decision it was too much trouble to use this feature to protect the Windows operating system, and that decision IMHO was the root of the vulnerability troubles that plague Windows today. I realize memory was precious in those days, but given that buffer overruns are the prime cause of malware trouble, just think of the grief and expense that could have been spared.
It is interesting to hear about the lengths people go to protect information about this thing, but what is it?
Not funny Ha Ha, but funny peculiar I guess. If the world's strangeness continues, I expect the day will come that Apple will embrace the trusted computing initiative and their compilers will start generating a new proprietary IL pseudocode. As an independent developer, I have tried to keep my code portable so as to leverage off marketing opportunities that require shifting platforms. Each time I have to play what I call "Engineering Poker" and guess which toolset will benefit me the most, it causes me grief. In my opinion, Apple doesn't need to play lock-in games. Mac OS X (Unix) is a superior platform to Windows, and the Apple hardware has been worth the extra cost for the longevity and stability. I think Apple would win on a totally level playing field.
Now that Microsoft has coerced the majority of keyboard manufacturers to include a button just for Microsoft Windows, they decide to deprecate the feature leaving the distribution channels full of keyboards that have a useless button. I suppose about the time the channels clear of the startbutton'd keyboards, Microsoft will put the feature back in and the manufacturers will be sitting on a ton of keyboards without the revived button. It is clear that modifying the keyboard design of the qwerty keyboard to support a single sourced proprietary operating system was a decision that seemed ok at the time, but now is nothing but trouble. As if supporting nationalized keyboards is not troublesome enough, we now have the permutation with and without the start button.
I have had one each of the last few versions of the kindle, and I am a happy user. Now I have a paper-white unit and I am very content. I have several hundred books comprised of novels, and technical books about software engineering. Sure my iPhone has a kindle app, but the paper-white is so readable and does not suffer from needing to do too many things. As much as I like "rooting", I really don't feel any need to poke around inside the kindle firmware. I have a Mac Pro, a MacBook Pro, an iPad Air, and an iPhone5. Each has a role in my life. I don't want to program on my ereader. Relating only to it's capability, and not focusing on Amazon's business practices, I highly recommend the Kindle. The 3G wireless networking is free and you can purchase and download books while outside wifi zones. It is an extra bonus that the kindle app is available for most of my other devices. If I really have to, I can reference a technical book on my iPhone. Even if a kindle-killer did come along, I already own quite a vfew kindle books, and would want to stick with that for the time being.
Spreadsheets are great for managing moderate two dimensional tables of data. I sometimes do my monthly budget on one. TFA is about the difference between vertical and horizontal applications. For people that can visualize, the spreadsheet is a powerful tool for managing moderate tables of data, such as my personal monthly budget. But I wouldn't try to use it for my complete yearly financial activity. Vertical applications are for people that benefit from pre-written software whose design embodies the details that help users. You can use spreadsheets for taxes, but knowledge apps for tax prep do a great job.
It is not about a legend. Microsoft did absolutely rig Windows so it wouldn't run on DR-DOS. That has been proven in court. Right down to the code that makes that happen. But the fact is that Microsoft was a ruthless marketing company and IMHO adversely affected the computer software industry by their actions.
Now that we have the words, it is possible to create an application the automatically generates a press release using all of them. There should be thousands of permutations possible. It could even find a home like the fortune cookie on Unix.
I have been wondering for several decades what was going on with Apple because they just do not seem to want to pay fair money for engineers. They know we want our resumes to say we worked for Apple. The thing that gets me the most is that the cost of living in Cupertino and surrounding areas it VERY high, the average commute being 45+ minutes the last time I looked. When I interviewed there years ago, I was astounded at what they wanted to pay. Since about 1999, someone has been draining the money out of the software engineering profession. I just don't see how they expect to get or keep the top talent with a policy like that.
"if that's above the infinite-bug-threshold, then you might as well not bother fixing any particular bug at that level, because the attacker can always just find another one. It doesn't even matter whether you have a prize program or not; the product is in a permanent state of unfixable vulnerability." Ah we are talking about Windows now eh?